How to do it?
- 1.
Paste your text
Paste or type text into the box - an article, notes, an essay or an email. The estimated listening time and the number of sentences appear below.
- 2.
Choose a voice
Pick a voice from the list; voices for your language are grouped first and the first English voice on your system is selected by default.
- 3.
Set the pace
Adjust speed (0.5-2×), pitch (0-2) and volume (0-100%) with the sliders. Changes apply from the next sentence.
- 4.
Listen and follow
Click “Read aloud”. The current sentence and word are highlighted; click any sentence to jump there, or use Pause, Resume and Stop.
Where the voices come from
The tool uses the browser’s Web Speech API. Voices are provided by your operating system or browser, not by our server, so the list differs between devices:
| Platform | Voices | How to add more |
|---|---|---|
| Windows 10/11 | Microsoft voices | Settings → Time & language → Speech → Add voices |
| macOS | system voices incl. enhanced ones | Accessibility → Spoken Content → System voice → Manage voices |
| iPhone/iPad | Siri-style voices | Accessibility → Spoken Content → Voices |
| Android | Google text-to-speech | Text-to-speech output → Google → Install voice data |
| Chrome, Edge | extra online voices | included |
| Linux | speech-dispatcher | install espeak-ng or RHVoice |
Voices marked “online” send the text to the browser vendor’s service; all other voices work offline on your device. Enhanced or “natural” voices sound noticeably better than compact default ones, so installing one is the single biggest quality improvement.
Long texts and reading time
Text is split into sentences (and long sentences at commas) and each piece is spoken separately. This avoids a known Chrome bug that cuts speech off after about 15 seconds, and enables sentence highlighting and jumping. There is no character limit.
At 1× a voice reads about 150 words per minute, so a 1,500-word article takes about 10 minutes, or about 6 minutes 40 seconds at 1.5×. Word highlighting depends on the voice - some online voices do not report word positions, so the whole sentence is highlighted instead. Pausing and resuming restarts the current sentence, which keeps playback reliable across browsers.
Practical uses
- Proofreading - hearing your own text exposes missing words, repetitions and clumsy sentences that the eye skips. Speed 0.9-1× works best.
- Language learning - choose a voice in the target language and lower the speed to 0.7× to hear pronunciation clearly.
- Accessibility and fatigue - listen to long documents instead of reading them on screen, useful for dyslexia or tired eyes.
- Presentations - check how long a script takes to say: the estimated time is shown before you press play.
Want to know the word count first? Paste the text into the character counter.
Can I download an MP3?
No. The Web Speech API plays speech directly through the system and does not give the page access to the audio, so no page built on it can honestly offer an MP3 download. To get a file, record system audio with a tool such as OBS Studio, use your operating system’s own export feature where one exists, or use a paid cloud text-to-speech service.
Frequently asked questions
Is it free?
+
Why does the voice have a foreign accent?
+
Can I save the speech as MP3?
+
Is my text sent anywhere?
+
Why does reading stop halfway?
+
How do I make the voice sound more natural?
+
Why is the voice list empty?
+
Updated: